ASUS joined GTC 2025 as a Diamond sponsor to showcase its latest AI POD with the NVIDIA GB300 NVL72 platform. The company has already received large orders, marking a major milestone. It is also showing its newest AI servers in the Blackwell and HGX lineup, which include several models with different NVIDIA technologies.
ASUS aims to help industries adopt AI by providing complete infrastructure solutions with NVIDIA platforms, helping clients reach markets faster. The ASUS AI POD uses the NVIDIA GB300 NVL72 server platform, giving businesses amazing processing power to handle huge AI challenges easily. Built with NVIDIA Blackwell Ultra, GB300 NVL72 leads AI innovation with better computing, more memory, and faster networking.
It has 72 NVIDIA Blackwell Ultra GPUs and 36 Grace CPUs in a rack design, delivering more AI FLOPs with up to 40 TB of high-speed memory per rack. It includes advanced networking, easy-to-service modules, 100% liquid cooling, and supports trillion-parameter LLM inference and training. This represents a big step forward in AI infrastructure, giving customers a reliable, scalable solution.
ASUS has proven it can build NVIDIA GB200 NVL72 infrastructure from scratch. It also shows the ASUS RS501A-E12-RS12U, a powerful software-defined storage server that cuts data training and inferencing latency. ASUS offers comprehensive services from hardware to cloud apps, covering design, cooling solutions, installation, deployment, and AI platforms.
Kaustubh Sanghani from NVIDIA said NVIDIA is working with ASUS to drive data center innovation. He mentioned that ASUS servers with the Blackwell Ultra platform will speed up training and inference, helping businesses unlock new possibilities in AI reasoning and agentic AI.
At GTC 2025, ASUS is also showing NVIDIA-certified servers that support apps built with NVIDIA AI Enterprise and Omniverse. These versatile models allow smooth data processing and computation, improving generative AI performance. The ASUS 10U ESC NB8-E11 has the NVIDIA Blackwell HGX B200 8-GPU for top AI performance.
The ASUS XA NB3I-E12 features HGX B300 NVL16 with more AI FLOPS, 2.3 TB of HBM3e memory, and advanced networking. Blackwell Ultra delivers great performance for AI reasoning, agentic AI, and video inference to meet changing data center needs. The 7U ASUS ESC N8-E11V dual-socket server has eight NVIDIA H200 GPUs and supports air or liquid cooling.
ASUS also offers servers for AI inferencing—the ASUS ESC8000 series with the latest NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. The ASUS ESC8000-E12P is a high-density 4U server for eight dual-slot NVIDIA H200 GPUs. It works with NVIDIA MGX architecture for flexible scaling and fast deployment. The ASUS ESC8000A-E13P supports eight dual-slot NVIDIA H200 GPUs and offers smooth integration for modern data centers.
ASUS also announced its new AI supercomputer, the ASUS Ascent GX10, in a compact package. Powered by the NVIDIA GB10 Grace Blackwell Superchip, it delivers 1,000 AI TOPS performance. The Ascent GX10 has a Blackwell GPU, a 20-core Arm CPU, and 128 GB of memory, supporting AI models with up to 200 billion parameters. This puts a petaflop-scale AI supercomputer right on the desks of developers and researchers worldwide.
Edge AI computers from ASUS IoT are also at GTC, showing their advanced AI inference capabilities. The ASUS IoT PE2100N edge AI computer delivers up to 275 TOPS, perfect for generative AI and language models. These enable natural language interactions with videos and images for event identification or object detection. With low power use and optional 4G/5G support, it works well for smart cities and robotics.
The ASUS IoT PE8000G rugged edge AI GPU computer supports dual 450 W NVIDIA RTX PCIe graphics cards and excels at real-time AI inference. With flexible power input and stability from -20°C to 60°C, it suits computer vision, autonomous vehicles, and video analytics in harsh environments. ASUS focuses on delivering excellence through innovative AI Infrastructure solutions for high-performance computing and AI workloads.
Their carefully designed servers and storage systems improve performance, reliability, and scalability for various enterprise needs. Their L12-ready infrastructure solutions use sustainable practices throughout the data center lifecycle, with energy-efficient power supplies and advanced cooling. Choosing an ASUS server goes beyond hardware, with software platforms for deployment and monitoring.
With 30 years of server industry experience since 1995, ASUS helps clients navigate the competitive AI landscape. Their AI-driven solutions enable dynamic workload management, allocating tasks to the most efficient resources. ASUS supports organizations as they handle AI infrastructure demands and drive innovation and sustainability.
ASUS aims to help industries adopt AI by providing complete infrastructure solutions with NVIDIA platforms, helping clients reach markets faster. The ASUS AI POD uses the NVIDIA GB300 NVL72 server platform, giving businesses amazing processing power to handle huge AI challenges easily. Built with NVIDIA Blackwell Ultra, GB300 NVL72 leads AI innovation with better computing, more memory, and faster networking.
It has 72 NVIDIA Blackwell Ultra GPUs and 36 Grace CPUs in a rack design, delivering more AI FLOPs with up to 40 TB of high-speed memory per rack. It includes advanced networking, easy-to-service modules, 100% liquid cooling, and supports trillion-parameter LLM inference and training. This represents a big step forward in AI infrastructure, giving customers a reliable, scalable solution.
ASUS has proven it can build NVIDIA GB200 NVL72 infrastructure from scratch. It also shows the ASUS RS501A-E12-RS12U, a powerful software-defined storage server that cuts data training and inferencing latency. ASUS offers comprehensive services from hardware to cloud apps, covering design, cooling solutions, installation, deployment, and AI platforms.
Kaustubh Sanghani from NVIDIA said NVIDIA is working with ASUS to drive data center innovation. He mentioned that ASUS servers with the Blackwell Ultra platform will speed up training and inference, helping businesses unlock new possibilities in AI reasoning and agentic AI.
At GTC 2025, ASUS is also showing NVIDIA-certified servers that support apps built with NVIDIA AI Enterprise and Omniverse. These versatile models allow smooth data processing and computation, improving generative AI performance. The ASUS 10U ESC NB8-E11 has the NVIDIA Blackwell HGX B200 8-GPU for top AI performance.
The ASUS XA NB3I-E12 features HGX B300 NVL16 with more AI FLOPS, 2.3 TB of HBM3e memory, and advanced networking. Blackwell Ultra delivers great performance for AI reasoning, agentic AI, and video inference to meet changing data center needs. The 7U ASUS ESC N8-E11V dual-socket server has eight NVIDIA H200 GPUs and supports air or liquid cooling.
ASUS also offers servers for AI inferencing—the ASUS ESC8000 series with the latest NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. The ASUS ESC8000-E12P is a high-density 4U server for eight dual-slot NVIDIA H200 GPUs. It works with NVIDIA MGX architecture for flexible scaling and fast deployment. The ASUS ESC8000A-E13P supports eight dual-slot NVIDIA H200 GPUs and offers smooth integration for modern data centers.
ASUS also announced its new AI supercomputer, the ASUS Ascent GX10, in a compact package. Powered by the NVIDIA GB10 Grace Blackwell Superchip, it delivers 1,000 AI TOPS performance. The Ascent GX10 has a Blackwell GPU, a 20-core Arm CPU, and 128 GB of memory, supporting AI models with up to 200 billion parameters. This puts a petaflop-scale AI supercomputer right on the desks of developers and researchers worldwide.
Edge AI computers from ASUS IoT are also at GTC, showing their advanced AI inference capabilities. The ASUS IoT PE2100N edge AI computer delivers up to 275 TOPS, perfect for generative AI and language models. These enable natural language interactions with videos and images for event identification or object detection. With low power use and optional 4G/5G support, it works well for smart cities and robotics.
The ASUS IoT PE8000G rugged edge AI GPU computer supports dual 450 W NVIDIA RTX PCIe graphics cards and excels at real-time AI inference. With flexible power input and stability from -20°C to 60°C, it suits computer vision, autonomous vehicles, and video analytics in harsh environments. ASUS focuses on delivering excellence through innovative AI Infrastructure solutions for high-performance computing and AI workloads.
Their carefully designed servers and storage systems improve performance, reliability, and scalability for various enterprise needs. Their L12-ready infrastructure solutions use sustainable practices throughout the data center lifecycle, with energy-efficient power supplies and advanced cooling. Choosing an ASUS server goes beyond hardware, with software platforms for deployment and monitoring.
With 30 years of server industry experience since 1995, ASUS helps clients navigate the competitive AI landscape. Their AI-driven solutions enable dynamic workload management, allocating tasks to the most efficient resources. ASUS supports organizations as they handle AI infrastructure demands and drive innovation and sustainability.